Answer
A solar pyranometer measures global solar irradiance in watts per square meter. IEC 61724-1 Class A monitoring needs ISO 9060 Class A pyranometers for both plane-of-array and global horizontal irradiance, calibrated every 2 years and cleaned weekly unless proven unnecessary. Mount the GHI sensor horizontal and the POA sensor parallel to the modules.
A solar pyranometer is the sensor behind most PV performance numbers. It measures how much sunlight reaches a surface, and every performance ratio, warranty check and yield comparison that follows inherits its errors. Selection, placement and upkeep matter as much as the label on the box.
Last updated: November 2026. This guide explains what a pyranometer measures, how ISO 9060 classes and IEC 61724-1 monitoring classes fit together, and how to mount, wire and maintain the sensors. Numbers come from named standards bodies, manufacturers and NREL. Where no published figure exists, the post says so.
Check modeled irradiance against your site data
Model the array and its losses, then compare the result with measured plane-of-array readings once the site is live.
Book a DemoChecking cost first? See the SurgePV pricing page.
What a Solar Pyranometer Measures
A solar pyranometer measures irradiance, the instantaneous power of sunlight per square meter, in W/m². The sensor has a 180-degree field of view, so it records direct beam radiation and diffuse radiation scattered by the sky.
The most common reading is global horizontal irradiance, or GHI, from a sensor mounted level. GHI is the usual starting point for yield models and site comparisons. For the full set of terms, see the GHI, DNI and DHI irradiance guide and the GHI glossary entry.
A pyranometer mounted parallel to the modules measures plane-of-array irradiance, or POA. It is what the modules actually receive, and it is the right comparison for production data. See the plane-of-array irradiance glossary entry for the definition.
Other arrangements include a shaded pyranometer for diffuse horizontal irradiance and a pair of pyranometers, one up and one down, to measure albedo. Seven Sensor notes that reflected and rear-side irradiance can be measured with inverted pyranometers on the array plane, or modeled from reflected irradiance.
Irradiance is power per area. Energy is its time integral, in kWh/m². Irradiance is the speedometer, energy is the odometer.
Pro Tip
Write the measurement geometry down before you choose a sensor: GHI level, POA parallel to the modules, rear-side facing down on bifacial arrays. A premium sensor in the wrong plane gives you precise data about the wrong thing.
Thermopile vs Silicon-Cell Pyranometers
Thermopile and silicon-cell sensors work differently, and only one family is covered by ISO 9060 classes.
A thermopile pyranometer uses a blackened detector under a glass dome. The detector heats in sunlight and the temperature difference produces a voltage. The OTT HydroMet specification guide lists the spectral range of its Class A thermopile pyranometer as 285 to 2800 nm, taken at the 50% points.
A silicon-cell pyranometer uses a photodiode or small PV cell and responds to a narrower band. It is quick, but the spectral mismatch can add error under cloudy skies, at low sun angles and with heavy aerosols.
| Feature | Thermopile pyranometer | Silicon-cell pyranometer |
|---|---|---|
| Detector | Blackened thermopile under a dome | Photodiode or small PV cell |
| Spectral response | Broad (285 to 2800 nm in the OTT guide’s Class A example) | Narrower band |
| ISO 9060 class | Class A, B or C | Check the datasheet |
| Response | Slower | Fast |
| Typical use | Bankable performance data | Trend checks, internal monitoring |
Thermopile spectral range from the OTT HydroMet Solar Weather Stations Specification Guide (2025). The silicon-cell column describes general characteristics, so read each manufacturer’s datasheet.
Key Takeaway
If the data will go to a lender, investor or warranty counterparty, use a classified thermopile instrument and keep its calibration certificate. Use silicon-cell sensors for internal checks where traceability is not required.
ISO 9060 Classes and IEC 61724-1 Monitoring
ISO 9060 classifies the pyranometer itself. IEC 61724-1 classifies the whole monitoring system. They are separate labels.
The OTT guide explains that ISO 9060:2018 defines three classes, A, B and C, with Class A the most accurate, and that Class A used to be called “secondary standard” in older versions. It defines a spectrally flat pyranometer as one with no more than 3% deviation in responsivity between 350 and 1500 nm, and a fast-response pyranometer as one with a response time under 0.5 seconds.
On the system side, the OTT guide describes an IEC 61724-1 Class A system as one designed for utility-scale and large commercial PV, typically over 1 MW. Hukseflux summarizes the Class A requirements as follows.
| Item | IEC 61724-1 Class A, per Hukseflux |
|---|---|
| Irradiance sensors | POA and GHI mandatory; diffuse and reflected or rear-side optional |
| Pyranometer class | ISO 9060 Class A with dew and frost mitigation, unless very little dew and frost is expected |
| Calibration | Every 2 years |
| Cleaning | Every week, unless it can be proven not necessary |
Source: Hukseflux, “How to calculate PV performance ratio and performance index”, checked November 2026.
This post does not reproduce IEC 61724-1 Class B requirements. They are not published on the pages checked, so read the standard itself before you write Class B into a contract.
Pro Tip
Put the monitoring class in the EPC contract before procurement. If it says Class A, the supplier must deliver the matching sensors, logger, calibration certificates and a cleaning plan, not just a Class A pyranometer.
Designing Pyranometer Placement
A pyranometer must see an unobstructed sky and sit in the plane it is meant to represent.
GHI sensor placement
The OTT HydroMet guide specifies installing the GHI pyranometer at least 2 meters above finished ground level, with no shadow from surrounding objects at any time. It achieves that by placing the sensor at a distance of at least 10 times the height of any nearby object. A 3 m structure, for example, would call for at least 30 m of clearance.
The same guide lists a bubble level accuracy better than plus or minus 0.1 degrees for its mounting hardware. Recheck level during maintenance visits, because mounts can shift.
Keep the sensor clear of reflective surfaces that could bounce light onto the dome.
POA sensor placement
Seven Sensor states that the sensor surface should be parallel to the plane it represents. For fixed-tilt arrays, that means matching module tilt and azimuth. The OTT guide specifies at least one Class A pyranometer for every fixed angle orientation of the modules.
On tracker sites, the sensor should move with the tracker so it sees the same plane as the modules. A fixed sensor plus a calculated angle cannot capture how a real tracker behaves.
Bifacial arrays
The OTT guide specifies three ISO 9060 Class B pyranometers for in-plane rear-side irradiance on each met station at a bifacial plant, mounted in the array and facing away from the front of the modules. Without a rear-side measurement you cannot check the bifacial gain assumed in the design.
Key Takeaway
Treat placement as part of the layout, not an afterthought. Reserve a level, unshaded spot for the GHI sensor and a representative position in the array for POA and rear-side sensors before the racking is ordered.
Outputs, Data Loggers and Wiring
Pyranometers give either analog or digital output. Analog sensors give a millivolt or current signal, and the logger needs the calibration factor from the certificate, usually in microvolts per W/m². Digital sensors send a calibrated value over a serial link.
The OTT guide lists Modbus RTU as the serial output for its pyranometers and a Modbus TCP/IP connection from the data logger to the plant SCADA network. If you are specifying a new installation, ask for Modbus RTU output so you avoid extra signal converters, and confirm the logger supports it.
Use shielded cable for analog signals, keep runs short, route them away from AC conductors, and label both cable ends with the sensor serial number so a later recalibration can be traced to the right channel.
Calibration and Maintenance
Pyranometers drift, domes age and soiling builds up, so schedule the work.
Hukseflux states that pyranometers in IEC 61724-1 Class A monitoring should be calibrated every 2 years and cleaned every week unless it can be proven this is not necessary.
NREL’s BORCAL, the Broadband Outdoor Radiometer Calibration, calibrates outdoors using the component summation method: a pyrheliometer traceable to the World Radiometric Reference measures the beam component, and the diffuse component is added to build the reference global irradiance, according to Hukseflux. An NREL study found that different calibration methods, indoor and outdoor, produced differences of +1% to +2% in irradiance measurements. Ask which method your certificate used.
A simple maintenance routine:
- Clean the dome at the interval set by your monitoring class and soiling rate.
- Check the desiccant window for moisture or fogging.
- Verify level at each site visit.
- Compare against a freshly calibrated reference when you can.
- Recalibrate on schedule and update the logger sensitivity factor straight away.
Key Takeaway
A skipped calibration or cleaning makes later performance ratio numbers hard to defend. Budget for both from year one, and keep the certificates with the site records.
Turn measured irradiance into a better design
Use the energy-yield model to test losses and shading, then compare predicted and measured production.
Book a DemoChecking cost first? See the SurgePV pricing page.
Common Pyranometer Mistakes
Mounting a fixed POA sensor on a tracker site
A tracker does not always sit at its modeled angle, because of stow logic, backtracking and mechanical tolerances. Mount the sensor on a tracker unit so it follows the same plane as the modules.
Using GHI where POA belongs in performance ratio
The traditional performance ratio is defined with in-plane irradiation, per Hukseflux. Using horizontal irradiance describes a different plane, so the comparison is not like for like. Use a co-planar sensor or a properly transposed value. The performance ratio calculation guide walks through the formula.
Assuming every Class A sensor behaves the same
ISO 9060 sets limits that every instrument in a class must meet, but not every model performs identically inside those limits. Specify the model as well as the class, and ask for calibration history.
Ignoring dirt
A dirty dome reads low. If a POA sensor reads low, the performance ratio can look better than the plant really is. Follow the cleaning interval and keep a record.
Skipping dew and frost mitigation
IEC 61724-1 Class A calls for dew and frost mitigation unless very little is expected at the site, per Hukseflux. Heated or ventilated sensors avoid data gaps at the start of the production day.
Using Irradiance Data in Design
At design stage you rarely have a year of site data. SurgePV’s energy-yield modeling and shading analysis let you test an array layout and its losses before the site is measured, and the generation and financial tool carries the result into financial modeling. When a pyranometer is later installed, compare its POA readings with the model to see whether the design assumptions held.
For layout work, see solar design software and shadow analysis. For the wider monitoring stack, read the guide to solar weather monitoring tools.
Choosing a Class by Project Type
| Project type | Starting point | Check |
|---|---|---|
| Utility-scale or large commercial, typically over 1 MW | IEC 61724-1 Class A system with ISO 9060 Class A pyranometers | OTT guide describes Class A systems as typical above 1 MW |
| Bifacial plant | Add rear-side POA sensors | OTT guide specifies three Class B sensors per met station |
| Smaller commercial site | Check what your contract or lender requires | Class B requirements are not reproduced here |
| Residential portfolio | Decide whether on-site sensors are needed at all | Cost, contract and reporting needs |
Prices vary by manufacturer, region and package, and this post does not publish a price table. Ask suppliers for current quotes for the sensors, logger, cabling, calibration and cleaning plan.
Conclusion
The right class, a correct mounting plane and a calibration and cleaning routine turn irradiance data into something you can defend in a performance review.
Three actions to take next:
- Match the class to the contract. Say Class A or Class B in writing, for the sensors and the system.
- Design placement into the layout. Reserve spots for GHI, POA and rear-side sensors.
- Fund calibration and cleaning from day one.
To compare modeled and measured production on your own projects, book a SurgePV demo.
Frequently Asked Questions
What is a solar pyranometer and how does it work?
A solar pyranometer is a radiometer that measures global solar irradiance on a flat surface, reported in watts per square meter. A thermopile sensor under a glass dome absorbs shortwave radiation and produces a voltage signal proportional to irradiance. The sensor sees a 180-degree hemisphere, so it captures both direct beam and diffuse sky radiation.
What is the difference between a thermopile and a silicon-cell pyranometer?
A thermopile pyranometer uses a blackened absorber that responds across the solar spectrum, and ISO 9060 Class A, B and C instruments are built this way. A silicon-cell pyranometer uses a photodiode that responds to a narrower band of wavelengths. Silicon-cell models are cheaper and faster, but the spectral mismatch can add error under cloudy skies and low sun angles.
What ISO 9060 class do I need for my solar project?
Use Class A where the data will be shown to lenders or tied to a performance guarantee, since IEC 61724-1 Class A monitoring calls for Class A pyranometers. Class B can suit smaller commercial monitoring if your contract allows it. Write the class into the contract so the supplier cannot substitute.
How do you mount a pyranometer for PV monitoring?
Mount a GHI pyranometer horizontally at least 2 meters above ground, at a distance of at least 10 times the height of any nearby object. Mount a POA pyranometer so its sensor surface is parallel to the module plane. On tracker sites, the POA sensor should move with the array.
How often should a pyranometer be calibrated?
Hukseflux, a pyranometer manufacturer, states that under IEC 61724-1 Class A monitoring the pyranometers should be calibrated every 2 years. The same source says they should be cleaned every week unless it can be proven this is not necessary. Follow your contract and the manufacturer’s calibration certificate.
What is plane-of-array irradiance and why does it matter?
Plane-of-array irradiance, or POA, is the solar radiation striking the tilted surface of the PV modules. The traditional performance ratio is defined with in-plane irradiation, so a POA measurement is the matching input. A horizontal GHI reading describes a different plane.
Do different calibration methods give different readings?
Yes. An NREL study found that different calibration methods produced differences of +1% to +2% in solar irradiance measurements. For a bankable project, ask which method and reference the calibration certificate used.
Sources
- Hukseflux: how to calculate PV performance ratio and performance index, checked November 2026
- Hukseflux: method for pyranometer calibration, checked November 2026
- OTT HydroMet Solar Weather Stations Specification Guide (2025)
- Seven Sensor: pyranometers for GHI, POA, diffuse and reflected irradiance, checked November 2026
- NREL: radiometer calibration methods and resulting irradiance differences (Habte et al.)
Where this fits
This article is part of SurgePV's Solar Design hub, which works through the topic from first principles to the decisions a project team actually has to make.
